Vehicle Trajectory Reconstruction With Variable-Width Path Corridor
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for fitting a vehicle trajectory to a reference path are inefficient, error-prone, and not scalable to complex systems, particularly for vehicles with trailers, and do not account for dynamic conditions such as weather and road-surface variations.
Innovation Solution
A method that constrains the vehicle's position within a variable-width corridor around the reference path, using a penalty on corridor width as part of the optimization process, allowing flexible and efficient numerical solving, and includes corridor width as an optimization variable to balance path tracking and smoothness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a trajectory is fitted to a reference path using conventional methods, then the trajectory follows the reference path, but the method is time-consuming and error-prone due to manual recording and不准确 maneuvering
Solution Approach 1:
The system automatically generates feasible trajectories by computing optimal control inputs from recorded reference paths, eliminating the need for manual trajectory recording and verification. The automated optimization process self-corrects errors and generates precise trajectories without human intervention.
Solution Approach 2:
The method pre-processes recorded reference paths by smoothing them and computing feasible trajectories in advance, so that when the autonomous vehicle needs to follow a path, the trajectory generation is already completed or can be quickly computed, saving operational time.
2Adaptability or versatility
If multiple paths are recorded for different weather conditions and vehicle configurations, then the system adapts to various conditions, but the complexity of path management increases significantly
Solution Approach 1:
The trajectory optimization algorithm is designed to be universal and can handle different vehicle dynamics models, weather conditions, and path types through a single unified framework. By parameterizing vehicle dynamics and using general optimization techniques, the system adapts to various conditions without requiring separate path recordings for each scenario.
Solution Approach 2:
The system adapts to different vehicle configurations and weather conditions by changing the parameters in the vehicle dynamics model and optimization constraints rather than requiring different recorded paths. The optimization algorithm adjusts to varying parameters such as vehicle mass, friction coefficients, and dynamic characteristics.
3Reliability
If existing trajectory methods are used for vehicles with trailers, then the method fails to account for trailer path and complex dynamics
Solution Approach 1:
The optimization framework incorporates dynamic vehicle models that can represent complex systems with trailers, articulated vehicles, and varying degrees of freedom. The algorithm computes time-varying control inputs that satisfy dynamic constraints, making the method applicable to diverse vehicle types while maintaining trajectory feasibility.
4Productivity
If a trajectory optimization problem is formulated numerically, then efficient solving is possible, but the problem may be ill-conditioned and difficult to solve robustly
Solution Approach 1:
The optimization problem is transformed by changing parameters such as the time horizon discretization, constraint formulations, and objective function weighting to improve numerical conditioning. The algorithm adjusts problem parameters to ensure stable and robust numerical solutions while maintaining computational efficiency.
Data Source
AI summary
A method for generating a vehicle trajectory by optimizing a performance measure J. The trajectory may include a sequence of states x=(xk)k=1N of the vehicle. The optimization is subject to predefined vehicle dynamics xk+1=ƒ(xk, uk), where uk is a control input to the vehicle, and a condition that each position of the vehicle shall be close to a reference path Xr. The vehicle's position is constrained inside a variable-width corridor around the reference path. A quantity r controlling the width of the corridor is included as an additional optimization variable and the performance measure includes a penalty on the corridor width. To define the corridor, each point of the reference path may be associated with a pair of laterally spaced ellipses and requiring each vehicle position to be outside the ellipses.